rl and rsrl

These are competitors—both are general-purpose reinforcement learning frameworks in Rust that serve the same use case, with rsrl being more mature and widely adopted (4x the stars, 3x the downloads).

rl
44
Emerging
rsrl
42
Emerging
Maintenance 0/25
Adoption 11/25
Maturity 16/25
Community 17/25
Maintenance 0/25
Adoption 14/25
Maturity 16/25
Community 12/25
Stars: 50
Forks: 11
Downloads: 28
Commits (30d): 0
Language: Rust
License: MIT
Stars: 202
Forks: 15
Downloads: 83
Commits (30d): 0
Language: Rust
License: MIT
Stale 6m No Package No Dependents
Stale 6m No Package No Dependents

About rl

benbaarber/rl

A rust reinforcement learning library

About rsrl

tspooner/rsrl

A fast, safe and easy to use reinforcement learning framework in Rust.

Provides modular RL algorithm implementations (Q-Learning, policy gradients) with linear function approximators and basis projections (Fourier, RBF) for continuous state spaces. Built on `ndarray` with optional BLAS acceleration for efficient numerical computation. Designed around composable traits for agents, policies, and environments, enabling rapid prototyping of RL experiments.

Scores updated daily from GitHub, PyPI, and npm data. How scores work